This project covers a broad base of studies aimed at assessing the epidemiology of the majority of hormonally-related cancers. Major efforts are underway for breast cancer, endometrial cancer, ovarian cancer, and testicular cancer. We also have an active research program on prostate cancer, covered in a separate report (Z01 CP010180-02). Our efforts for all of these cancers relate to a variety of environmental, genetic and hormonal predictors of risk. A large multi-disciplinary study was conducted in Poland between 2000-2004 to assess risk factors for breast, ovarian and endometrial cancers. The study involved collection of multiple biologic samples, with a primary aim of assessing biomarker associations with risk and the interactive effects of genetic and environmental determinants of risk. In addition, special components of the study addressed effects of physical activity, occupational factors, and household chemical exposures. For physical activity, special efforts were expended to improve exposure assessment via having women wear accelerometers. The study also involved collection of tissue samples to enable histological and molecular tumor classification (e.g. utilizing tissue microarray techniques for immunohistochemistry analyses). The large amount of data collected has enabled a number of analyses. With respect to breast cancer, recent publications have evaluated effects of genetic polymorphisms and etiologic heterogeneity by tumor tissue markers. Breast cancer risk is also of major interest in a follow-up of a cohort of women previously screened for bone density. This resource, which previously involved collection of serologic samples and detailed questionnaire data, is enabling an assessment of the interrelationship of bone density, genetic factors and endogenous hormones in predicting subsequent cancer occurrence. It is well recognized that breast cancers that occur among Africans and African-Americans tend to exhibit different clinical characteristics as compared with Caucasians, incluidng a higher prevalence of estrogen receptor negative and triple negative tumors, cancers associated with a generally poor prognosis. To better understand the reasons for the occurrence of these cancers, we are developing plans to undertake a case-control study in Ghana, where incidence rates of breast cancer have been increasing. The study is being designed to evaluate some novel etiologic hypotheses as well as to relate risk factors to carefully defined subtypes of breast cancers. Mammographic density has been recognized as a major predictor of subsequent breast cancer risk, but the biologic basis for this association is unclear. Funds made available through the sale of breast cancer stamps supported a study to assess hormonal and immunologic correlates of mammographic density among a group of women receiving breast biopsies at the University of Vermont. A collaboration has been established with the Gynecologic Oncology Group to determine means of collecting epidemiologic data within the context of a number of ongoing trials. A standardized questionnaire has been developed and integrated into a large trial of endometrial cancer. This effort should be useful in assessing epidemiologic predictors and molecular markers associated with carefully defined histologic subgroups of tumors. We have learned much about the natural history of cervical cancer (as described in another project report) and are now anxious to expand our knowledge in this area to address the natural history of another gynecologic tumor, namely endometrial cancer. Endometrial hyperplasias are recognized to increase the subsequent risk of endometrial cancer, but data with which to accurately predict risk are lacking, and it is unknown how other factors might influence those risks. We have conducted a nested case-control study within a prepaid health plan to better understand the risk of endometrial cancer in women diagnosed with endometrial hyperplasia. Data from this study have supported that notion that atypical hyperplasia is strongly related to subsequent endometrial cancer risk. We are also in the pilot phases of developing a study to assess early markers which may be important to the development of ovarian cancer and endometrial cancer. Testicular cancer is the most common type of cancer among American men aged 15-35 years and occurs five times more frequently among white men than among black men. To study the etiology of this tumor, we have conducted a large case-control study of U.S. servicemen in collaboration with the U.S. Department of Defense (DoD). Analysis of pre-diagnostic serum samples has found that exposure to organochlorine pesticides (specifically, dichlorodiphenyldichloroethylene (DDE) and chlordane) is significantly associated with testicular cancer. Serum analyses have also found that men who develop testicular cancer have aberrant gonadotropin levels prior to diagnosis. Additional findings indicate that while adult height is associated with risk, age at puberty, consumption of dairy products and maternal smoking in pregnancy are not associated. In addition, genetic variability in the insulin-like growth factor pathway and the hormone metabolizing pathway does not appear to be related to risk. Genetic variability in the inhibin pathway, however, may be associated. As a complement to this research, other studies have being undertaken to identify underlying causes of cryptorchism, a recognized risk factor for testicular cancer. These studies have compared the prevalence of cryptorchism, as well as risk factors for the anomaly, between black and white populations. Though the prevalence of cryptorchism is significantly higher among white populations, the small difference in the rate is not compatible with the large difference in testicular cancer risk. Risk factors between the two groups do not vary greatly and differences in maternal hormone levels are not related to cryptorchism. This project has also included a focus on the etiologic role of endogenous hormones for a variety of tumor sites. In these studies, attention is focusing not only on classically accepted hormones, but also on some newly suggested predictors, including adiponectin levels. An extended follow-up of participants in the Columbia, Missouri component of the Mayo Serum Bank has been undertaken. We have used a newly developed technique at a laboratory in Frederick that uses liquid chromatography/mass spectometry to measure 16 estrogens and their metabolites. We are also measuring the estrogen panel in our BFIT followup study, which will allow us an opportunity to examine the interrelationships between estrogens and measured bone density as they relate to subsequent risk of breast, colorectal, endometrial and ovarian cancers. We are currently collaborating with investigators of the Womens Health Initiative to measure estrogens in relation to ovarian and endometrial cancers that developed among participants in the observational component of that investigation. We also have developed plans to measure estrogens and androgens in relation to male breast, testicular and esophageal cancers. Finally, in the Polish study we are measuring urinary estrogens among the control subjects in order to more fully understand relationships with identified risk factors, including physical activity levels that have been objectively determined.